Implementations of various RL algorithms using OpenAI Gym environments
- [REINFORCE] (https://github.com/justinjfu/chaos_theory/blob/master/chaos_theory/algorithm/reinforce.py)
- [DDPG] (https://github.com/justinjfu/chaos_theory/blob/master/chaos_theory/algorithm/ddpg_old.py)
- NAF
TODO:
- Cross-Entropy Method
- TRPO/Natural Policy Gradients
- QProp
- Python 2.7
- [OpenAI Gym] (https://github.com/openai/gym)
- [Tensorflow] (https://github.com/tensorflow/tensorflow)
- Numpy/Scipy
- imageio (for recording gifs)
REINFORCE Cartpole experiment:
python scripts/run_reinforce.py
DDPG Half-Cheetah experiment:
python scripts/run_ddpg2.py